Download Human Atherosclerosis and the Diet

Human
Atherosclerosis
and
the Diet
By ANCEL KEYS, PH.D.
Downloaded from http://circ.ahajournals.org/ by guest on August 3, 2017
D ISCUSSIONS about the possible effects of the diet on the development
of human atherosclerosis center on
five items: (1) calorie excess and the resulting
obesity, (2) cholesterol in the diet, (3) animal
fats in the diet, (4) total fats in the diet, (5)
the dietary supply of substances which may
have a "lipotropic" action, for example,
lecithin, choline, inositol. Since the data on the
possible action in man of the lipotropic substances pertain to pharmacologic rather than
dietary levels and, moreover, are either questionable or negative in regard to an effect on
atherosclerosis, the present discussion will
not make further mention of these substances.
There are abundant actuarial data which
show there is an elevated incidence of fatal
atherosclerosis in obese or overweight people.
Routine necropsy studies indicate a direct
relationship between atherosclerotic changes in
the arteries, including the deposition of cholesterol and lipids, and relative fatness of the
whole body. Experiments on the production of
atherosclerosis in animals likewise show that
the relative calorie intake is important.3 But
atherosclerosis occurs in many persons who
are neither fat nor overweight. Our first real
conclusion, then, is that the development of
the atherosclerotic process is abetted, at least
in some persons, by an excessive calorie intake
and/or the resulting obesity. The calorie
level of the diet is an influential but not a
controlling factor.
At present the greatest concern about the
diet in regard to atherosclerosis has to do with
the consumption of foods which contain
cholesterol. In some animals, notably the
rabbit and the chicken, the addition of large
amounts of cholesterol to the diet can lead to
high serum cholesterol concentrations and to
arterial lesions which resemble human atherosclerosis in many respects.
Unfortunately, these animal experiments do
not necessarily have any relevance to the ques-
tion of the role of the diet in human atherosclerosis. No animal species close to man in
metabolic habitus has been shown to be susceptible to the induction of atherosclerosis by
cholesterol feeding. The nearest approach to
metabolic comparability is the dog which requires extensive thyroid damage as well as
tremendous amounts of dietary cholesterol
before positive effects can be elicited. Moreover, even in the favorite species for such experimentation, the herbivorous rabbit, the
necessary concentration of cholesterol in the
diet is fantastically high in comparison with
actual human diets. The actual range of daily
intakes of cholesterol provided by real human
diets is from zero to perhaps 1000 mg., but the
vast majority of human diets seldom average
as much as 700 mg. The cholesterol levels in
the diets used to induce atherosclerosis in
animals range from 0.5 to 5 per cent by weight
of the dry food, the most popular level being
2 per cent. This means something like 1000
to 10,000 mg. of cholesterol per 1000 Calories
of food, the 2 per cent cholesterol level being
equivalent to about 4000 mg. per 1000 Calories
if 30 per cent of the calories are derived from
fats. We should have to provide some 10,000
to 15,000 mg. of cholesterol daily to a man to
be comparable. Moreover, there is reason to
believe that man has a greater power of
cholesterol regulation than does the rabbit or
the chicken. From the animal experiments
alone the most reasonable conclusion would
be that the cholesterol content of human diets
is unimportant in human atherosclerosis.
Direct evidence on man in this connection is
unimpressive. Arguments about the incidence
of atherosclerosis and the customary intake of
cholesterol in the Far East are based on no
real data on either atherosclerosis incidence
or the character of the diet. The recent analysis of mortality data in Scandinavia" is more
acceptable but still suffers from the inherent
weakness of attempting to discover causation
from a parallelism between crude estimates of
national averages for two variables. Besides
From the Laboratory of Physiological Hygiene,
University of Minnesota, Minneapolis, Minn.
115
Circulation, Volume V, January, 1952
116
CLINICAL PROGRESS
Downloaded from http://circ.ahajournals.org/ by guest on August 3, 2017
the dubious point of attributing war-time
changes in vital statistics to actual changes
in atherosclerosis, the analysis glosses over the
fact that the presumed variations in cholesterol intake were parallel to similar variations
in total dietary fat and, at least to some extent,
to changes in calories intake and physical
activity (see also13).
Such, in substance, is the actual evidence
on human atherosclerosis itself. But significant inferences can be made from indirect
evidence if we admit that there is an important relationship between atherosclerosis and
the concentration of cholesterol in the serum.
There is overwhelming evidence for such a
relationship:
(1) On the average, wherever the serum
cholesterol tends to be high-as in diabetes,
nephrosis, hypothyroidism, and idiopathic hypercholesterolemia-there is a tendency toward early and severe atherosclerosis.
(2) Persons with definite evidence for disease of the coronary arteries tend to have
elevated serum cholesterol values4 especially
when they are compared with properly matched
"normals."5 Good confirmation can be obtained from recent raw data. 6, figs. 14, 15
(3) There is at least some parallelism between the age trend of cholesterol concentration in the serum and the age trend in atherosclerosis development.7' 9
(4) In animal experiments there is a close
parallelism between the serum cholesterol
level and the development of experimental
"'atherosclerosis."
There is much new evidence on the effect
of the diet on the serum cholesterol level in
man. In the first place, the serum cholesterol
level of man is independent of large differences
in the habitual voluntary intake of cholesterol.7 8 Several important confirmations of this
fact have been published recently.4 '4 Colossal
intakes, however, far beyond those possible
in natural diet, may cause a slow rise in the
serum cholesterol.'2
The rice-fruit diet, which is devoid of both
fats and cholesterol, produces a prompt fall
in the serum of men with normal levels of
cholesterols and a dramatic fall in patients with
severe idiopathic hypercholesteremia.8 We
have found that the addition of vegetable fat
to this cholesterol-free diet is followed by an
equally dramatic rise of serum cholesterol to
previous levels.
We have been able recently to make systematic studies on these questions with men
under full supervision on diets rigidly controlled for some four months. Starting (and
ending) with standard diets providing daily
intakes of 150 Gm. of fat and 600 to 700 mg.
of cholesterol, the following findings emerged:
(1) Change to a diet devoid of cholesterol
and providing 15 Gm. of fat daily produced a
rapid fall in serum cholesterol to about 80
per cent of the control value.
(2) Almost identical results were obtained
with the same diet to which had been added
about 500 mg. of cholesterol daily.
(3) Change to a diet devoid of cholesterol
but providing about 70 Gm. of fat daily, produced a decline in serum cholesterol about half
as great as when the fat level was only 15 Gm.
(No. 1, above).
(4) The addition of 500 mg. of cholesterol
daily to the diet (No. 3, above) providing 70
Gm. of fat did not change the result.
It should be noted that all of these experimental diets were comparable in calories and
proteins, that the differences in fat intake were
achieved by changing the amount of vegetable
fat in the diet, and that the differences in
cholesterol intake were achieved by adding or
withholding whole egg yolks (from which
cholesterol is readily absorbed). Finally, with
each of the diets the major change in serum
cholesterol occurred in the first week or so,
and there was a tendency to reach a plateau,
or at least a much slower rate of change, in
subsequent weeks.
Much longer-term experiments on man are
desirable but, so far as they go, these data are
consistent within themselves and with all other
facts known to me. The ensemble points
strongly to the conclusion that, other things
being equal, the serum cholesterol level in
adult man is independent of the cholesterol
intake over the range of zero to at least 700
mg. daily. But the fat intake is quite another
ATHEROSCLEROSIS. A SYMPOSIUM
Downloaded from http://circ.ahajournals.org/ by guest on August 3, 2017
matter and appears to have great importance.
However, there is not the slightest evidence
for a difference between animal and vegetable
fat in this regard.'
It is now well recognized, and should have
been clear long ago, that all animals, including
man, have a large capacity for synthesizing
cholesterol. Our experiments on fat-free diets
also show that man has a large capacity to
remove the cholesterol already present in the
blood serum. On such diets the rate of removal
of cholesterol from the serum seems to be proportional to the initial (control) serum level of
cholesterol. In one patient with hypercholesteremia (900 mg. per 100 ml. of serum), the
rate of disappearance from the blood was about
1500 mg. daily for several weeks on a fat-free
diet. But with the same cholesterol-free diet
to which were added about 50 gm. of vegetable
fat. there was an increase of cholesterol in his
total serum amounting to about 1000 mg. a
day for a period of several weeks.
Now, clearly the cholesterol level is not the
whole story in the development of atherosclerosis; otherwise the patient mentioned above
would have been dead long ago. But, in so
far as there is an important relationship between serum cholesterol and atherosclerosis,
these dietary data have both theoretic and
practical implications of some consequence.
A final question may be raised about the significance of all this in view of the studies by
Gofman and his colleagues on the materials
separated in the ultracentrifuge. From the
evidence reported so far we may make two
comments and draw a tentative conclusion:
(1) The response to diet of the serum concentration of substances characterized by -10
to -20 Svedberg units of sedimentation (the
"G" substances) appears to be similar to our
findings with serum cholesterol, and
(2) There is as yet no reason to suggest that
the concentration of G substances in the serum
is any more closely related to atherosclerosis
than is the concentration of total cholesterol.
(3) At present, we may conclude that the
implications for atherosclerosis drawn from
dietary effects on serum cholesterol would be
117
similar if the serum criterion were the concentration of the G substances.
In summary, then, we may remark that
direct evidence on the effect of the diet on
human atherosclerosis is very little and is
likely to remain unsatisfactory for a long time.
But such evidence as there is, plus valid inferences from indirect evidence, suggests that a
substantial measure of control of the development of atherosclerosis in man may be achieved
by control of the intake of calories and of all
kinds of fats, with no special attention to the
cholesterol intake. This means: (1) avoidance of
obesity, with restriction of the body weight to
about that considered standard for height at
age 25; (2) avoidance of periodic gorging and
even temporary large calorie excesses; (3)
restriction of all fats to the point where the
total extractable fats in the diet are not over
about 25 to 30 per cent of the total calories;
(4) disregard of cholesterol intake except,
possibly, for a restriction to an intake less than
1 Gm. per week.
1 ABELIN,
REFERENCES
I.: Zur Frage der Sterinbildung im Tier-
k6rper. II. Mitteilung. Helvet. physiol. et pharmacol. Acta 6: 879, 1948.
2 CHAPMAN, C. B., GIBBONS, T., AND HENSCHEL, A.:
The effect of the rice-fruit diet on the composition of the body. New England J. Med. 243: 899,
1950.
3FIRSTBROOK, J. B.: Factors influencing the atherosclerotic process. Circulation 2: 464, 1950.
4 GEIER, F. M.: Cholesterol and coronary artery disease. Permanente Found. M. Bull. 7: 49, 1949.
5 GERTLER, M. M., GARN, S. M., AND WHITE, P. D.:
Diet, serum cholesterol and coronary artery disease. Circulation 2: 696, 1950.
6 GOFMAN, J. W., JONES, H. B., LINDGREN, F. T.,
LYON, T. P., ELLIOTT, H. A., AND STRISOWER,
B.: Blood lipids and human atherosclerosis.
Circulation 2: 161, 1950.
7 KEYS, A.: The physiology of the individual as an
approach to a more quantitative biology of man.
Federation Proc. 8: 523, 1949.
8 -, MICKELSEN, 0., MILLERX E. V. 0., AND CHAPMAN, C. B.: The relation in man between
cholesterol levels in the diet and in the blood.
Science 112: 79, 1950.
9 ,-, _,HAYES, E. R., AND TODD, R. L.: The
concentration of cholesterol in the blood serum
of normal man and its relation to age. J. Clin.
Investigation 29: 1347, 1950.
118
10
CLINICAL PROGRESS
MALMROS, H.: The relation of nutrition to health.
A statistical study of the effect of the war-time
on arteriosclerosis, cardiosclerosis, tuberculosis
and diabetes. Acta med. Scandinav. Suppl. 286:
137, 1950.
1 MELLINKOFF, S. M., MACHELLA, T. E., AND REINHOLD, J. G.: The effect of a fat-free diet in causing low serum cholesterol. Am. J. M. Sc. 220:
203, 1950.
12
MESSINGER, W. J., POROSOWSKA, Y., AND STEELE,
J. M.: Effect of feeding egg yolk and cholesterol
on serum cholesterol levels. Arch. Int. Med. 86:
189, 1950.
13 SCHETTLER, G.: Zum Einfluss der Ernahrung auf
den Cholesteringehalt des Blutes. Klin. Wchnschr. 28: 569, 1950.
14 WILKINSON C. F. JR., BLECHS, E., AND REIMER,
A.; Is there a relation between diet and blood
cholesterol? Arch. Int. Med. 85: 389, 1950.
Downloaded from http://circ.ahajournals.org/ by guest on August 3, 2017
Human Atherosclerosis and the Diet
ANCEL KEYS
Downloaded from http://circ.ahajournals.org/ by guest on August 3, 2017
Circulation. 1952;5:115-118
doi: 10.1161/01.CIR.5.1.115
Circulation is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
Copyright © 1952 American Heart Association, Inc. All rights reserved.
Print ISSN: 0009-7322. Online ISSN: 1524-4539
The online version of this article, along with updated information and services, is located on
the World Wide Web at:
http://circ.ahajournals.org/content/5/1/115.citation
Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally
published in Circulation can be obtained via RightsLink, a service of the Copyright Clearance Center, not
the Editorial Office. Once the online version of the published article for which permission is being
requested is located, click Request Permissions in the middle column of the Web page under Services.
Further information about this process is available in the Permissions and Rights Question and Answer
document.
Reprints: Information about reprints can be found online at:
http://www.lww.com/reprints
Subscriptions: Information about subscribing to Circulation is online at:
http://circ.ahajournals.org//subscriptions/